has anyone played with subtle amounts of anisotropic smoothing (3danisosmooth) to prep an image for FreeSurfer? if so does it make a difference, esp re: pial / dura separation? obviously i can try a bunch of parameters myself but first i figured i'd see if anyone has already tested this. specifically, i'm working with a (background noise-suppressed) 7T MP2RAGE but i hope the question is of general interest.
Hi, Sam-
This kind of image preprocessing is interesting to consider.
However, I haven't used anisotropic smoothing. I would worry about interactions with internal FreeSurfer processing and assumptions. I would ask the FS developers their opinions about preliminary processing.
--pt